Finnish men who saunaed 4–7 times a week had 63% fewer sudden cardiac deaths — a figure resting on 201 men and 10 deaths. Then 20 randomised trials measured what sauna actually changes in the body, and found almost nothing.
Nearly half a million people a month look up sleep hygiene. The clinical guideline advises clinicians against using it as a standalone treatment — and when 241 trials were pulled apart to weigh each ingredient, it scored 1.01. Nothing.
A trial put box breathing head to head against two other techniques and against meditation. All three breathing groups improved mood by almost the same amount — and only 21 people were doing the famous square.
Around 117,000 Japanese adults were followed for years, and people who said their life felt worth living died less. None of those studies used the four-circle diagram — which began life in 2012 with a different word in the middle: purpose.
A 2025 meta-analysis of 24 cohorts found 7,000 steps a day linked to 47% lower mortality, 38% less dementia and 22% fewer depressive symptoms — with the curve bending around 5,000 to 7,000. Cancer risk did not move. And your age changes where the number sits.
Across 34 double-blind trials, magnesium lowered blood pressure about 2 points — small, real, and the authors called it causal. Cochrane found nothing at all for leg cramps, at high certainty. And the entire sleep evidence base is three trials in 151 people.
A 2003 meta-analysis examined 250 supplements; only six had enough studies to analyse and only two worked. Creatine was one. Modern trials put it at about 2.5 lb of extra lean mass in under-50s — and 3 lb in older adults.
Cochrane pooled 86 trials and 162,796 people: no effect on dying or on cardiovascular events, at high certainty. The benefit that is real needs 4 grams a day. And omega-3 raised atrial fibrillation risk 25% across 81,210 patients.
A 790-man NIH trial and an individual-participant meta-analysis of 35 trials both find real improvement in sexual function and quality of life. The trial designed to test energy found nothing. And the only trial big enough to count fractures found 43% more of them.
Twelve men trained both arms for 15 weeks, one arm in a bath of exercise-induced testosterone and growth hormone. The hormone-free arm grew 12%. The hormone-bathed arm grew 10%. Replicated in 49 trained men and 56 more.